A remarkably facile zirconium(IV) → aluminum(III) β-diketiminate transmetalation that also results in a more active olefin polymerization catalyst upon activation

Article abstract of DOI:10.1021/om049364g

(R-IAN)₂ZrX₂ complexes, where X = NMe₂, Cl, have recently emerged as effective olefin polymerization precatalysts bearing chiral ligands formally of the β-diketimine class. An attempt to exchange dimethyl-amido ligands for alkyl (methyl) using the Jordan protocol resulted in a surprisingly facile transmetalation of the bidentate Me-IAN ligands from zirconium(IV) to aluminum(III). An initial study of the process and the finding that the (R-IAN)AlMe₂ complexes that result are more potent precatalysts themselves provide a case study within the rapidly growing area of catalysis based on β-diketiminate metal complexes.

Full text of DOI:10.1021/om049364g

Organometallics 2004, 23, 5885-5888
5885
A Remarkably Facile Zirconium(IV) f Aluminum(III)
â-Diketiminate Transmetalation That Also Results in a
More Active Olefin Polymerization Catalyst upon
Activation
Sarah B. Cortright,† Joseph N. Coalter III,‡ Maren Pink,† and
Jeffrey N. Johnston*,†
Department of Chemistry, Indiana University, Bloomington, Indiana 47405-7102, and
The Dow Chemical Company, 3200 Kanawha Terrace, South Charleston, West Virginia 25303
Received August 16, 2004
Summary: (R-IAN)2ZrX2 complexes, where X ) NMe2,
Cl, have recently emerged as effective olefin polymeri-
zation precatalysts bearing chiral ligands formally of the
â-diketimine class. An attempt to exchange dimethyl-
amido ligands for alkyl (methyl) using the Jordan
protocol resulted in a surprisingly facile transmetalation
of the bidentate Me-IAN ligands from zirconium(IV) to
aluminum(III). An initial study of the process and the
finding that the (R-IAN)AlMe2 complexes that result are
more potent precatalysts themselves provide a case study
within the rapidly growing area of catalysis based on
â-diketiminate metal complexes.
IAN amines 1 (amine ligands formed from isoquino-
line and 2-aminonaphthalene subunits) are bidentate
N,N ligands formally of the â-diketimine class but are
more appropriately identified as â-amido Schiff bases
on the basis of their binding characteristics.11 As part
The development of â-diketimines as supporting
ligands for metal-centered reactions constitutes one of
the most active research areas in recent years. This
phenomenon has been largely driven by their promise
as complements, if not surrogates, for metallocene
catalysts1 and their use as electron-rich bidentate
ligands that can be rendered substantially more hin-
dered than their 1,3-dicarbonyl counterparts.2 Through
the efforts of numerous investigators, general guidelines
have been drawn for the formation of group IV metal
â-diketiminate complexes and their ability to polymerize
R-olefins.3-6 Moreover, the challenges that face this
ligand motif and their solutions have brought an
increased diversity to the field of Ziegler-Natta
polymerization.7-10
of a general program to develop the asymmetric cataly-
sis of chiral â-diketimine-supported metal complexes,12
we have studied their unique coordination chemistry
and have systematically evaluated the activity of these
complexes in ethylene polymerization as a function of
catalyst structure and the ancillary ligand removed
to form the putative cationic polymerization cata-
lyst.13,14 Accordingly, the low solubility of (R-IAN)2ZrX2
dihalides (X ) Cl, I) led us to attempt synthesis of
their dialkyl derivatives. Whereas exposure of R-IAN
amines of various structure (R ) Me, Ph) to ZrBn4
readily provided the 1:1 complexes (R-IAN)ZrBn3,
the corresponding 2:1 complexes (R-IAN)2ZrBn2 of po-
tential C2 symmetry were not forthcoming by this
approach.
* To whom correspondence should be addressed. E-mail: jnjohnst@
indiana.edu.
† Indiana University.
We therefore turned to the Jordan amido/alkyl ex-
change protocol using AlMe3, a technique primarily
‡ The Dow Chemical Company.
(1) Review of â-diketimine coordination chemistry: Bourget-Merle,
L.; Lappert, M. F.; Severn, J. R. Chem. Rev. 2002, 102, 3031.
(2) (a) Kakaliou, L.; Scanlon, W. J. IV; Qian, B.; Baek, S. W.; Smith,
M. R., III; Motry, D. H. Inorg. Chem. 1999, 38, 5964. (b) Basuli, F.;
Bailey, B. C.; Brown, D.; Tomaszewski, J.; Huffman, J. C.; Baik, M.-
H.; Mindiola, D. J. J. Am. Chem. Soc. 2004, 126, 10506. (c) Eckert, N.
A.; Smith, J. M.; Lachicotte, R. J.; Holland, P. L. Inorg. Chem. 2004,
43, 3306.
(9) Vollmerhaus, R.; Rahim, M.; Tomaszewski, R.; Xin, S.; Taylor,
N. J.; Collins, S. Organometallics 2000, 19, 2161.
(10) Jin, X.; Novak, B. M. Macromolecules 2000, 33, 6205.
(11) Cortright, S. B.; Yoder, R. A.; Johnston, J. N. Heterocycles 2004,
62, 223.
(12) Enantioselective chiral â-diketimine-supported metal cataly-
sis: (a) Leutenegger, U.; Madin, A.; Pfaltz, A. Angew. Chem. 1989,
101, 61. (b) Fritischi, H.; Leutenegger, U.; Pfaltz, A. Angew. Chem.
1986, 98, 1028. (c) Pfaltz, A. In Enantioselective Catalysis With Chiral
Cobalt And Copper Complexes; Scheffold, R., Ed.; Modern Synthetic
Methods Vol. 5; Springer-Verlag: Berlin, Heidelberg, 1989. (d) Fritis-
chi, H.; Leutenegger, U.; Pfaltz, A. Helv. Chim. Acta 1988, 71, 1553.
(e) Lo, M.-C. L.; Fu, G. C. J. Am. Chem. Soc. 2002, 124, 4572.
(13) (a) Cortright, S. B.; Huffman, J. C.; Yoder, R. A.; Coalter, J.
N., III; Johnston, J. N. Organometallics 2004, 23, 2238. (b) Cortright,
S. B.; Johnston, J. N. Angew. Chem., Int. Ed. 2002, 41, 345.
(14) For similar findings, see: Kettunen, M.; Vedder, C.; Schaper,
F.; Leskela¨, M.; Mutikainen, I.; Brintzinger, H.-H. Organometallics
2004, 23, 3800.
(3) Rahim, M.; Taylor, N. J.; Xin, S.; Collins, S. Organometallics
1998, 17, 1315.
(4) Deelman, B.; Hitchcock, P. B.; Lappert, M. F.; Leung, W.; Lee,
H.; Mak, T. C. W. Organometallics 1999, 18, 1444.
(5) Giannini, L.; Solari, E.; De Angelis, S.; Ward, T. R.; Floriani, C.;
Chiesi-Villa, A.; Rizzoli, C. J. Am. Chem. Soc. 1995, 117, 5801.
(6) Martin, A.; Uhrhammer, R.; Gardner, T. G.; Jordan, R. F.;
Rogers, R. D. Organometallics 1998, 17, 382.
(7) Reviews: (a) Britovsek, G. J. P.; Gibson, V. C.; Wass, D. F.
Angew. Chem., Int. Ed. Engl. 1999, 38, 428. (b) Piers, W. E.; Emslie,
D. J. H. Coord. Chem. Rev. 2002, 233, 131.
(8) Qian, B.; Scanlon, W. J., IV; Smith, M. R., III; Motry, D. H.
Organometallics 1999, 18, 1693.
10.1021/om049364g CCC: $27.50 © 2004 American Chemical Society
Publication on Web 11/09/2004

5886 Organometallics, Vol. 23, No. 25, 2004
Communications
Scheme 1. Amide/Alkyl Exchange and Transmetalation Pathways from (R-IAN)2Zr(NMe2)2
utilized with complexes of the metallocene type.15,16 For
complexes using bidentate nitrogen-based ligands, suc-
cessful exchange reactions of this type have been
documented for group IV metals.17,18 Our attempt to
directly convert 2a to the dimethyl derivatives followed
the typical protocol in which (Me-IAN)2Zr(NMe2)2 (2a)
in benzene was treated with 4 equiv of AlMe3 (Scheme
1, eq 1). The deep red solution of 2a immediately
lightened to yellow and over a 30 min period darkened
again to a deep red-violet. Removal of the volatile
components by vacuum led to a red-black oil that was
redissolved in d6-benzene. The relative simplicity of the
aromatic region of its 1H NMR spectrum confirmed that
a single ligand environment existed, indicating that
either the C2 symmetry had been preserved or a 1:1
ligand/metal complex was present. However, the methyl-
(metal) resonances appeared as two magnetically in-
equivalent singletssan observation clearly inconsistent
with the desired C2-symmetric dimethyl product 4a.
A second reaction between (Me-IAN)2Zr(NMe2)2 (2a)
and AlMe3 in which the latter was limited to 2 equiv
provided a similar result. After 90 min at 25 °C, the
volatile components were again removed by vacuum to
reveal only partial conversion to the previously observed
product. Another aromatic product and several broad
unidentifiable peaks were also noted; however, no 2a
remained. When this solution was subjected to heating
for 2 h at 75 °C, 1H NMR revealed complete conversion
to the previously observed complex. Lowering of the
relative amount of AlMe3 below 2 equiv consistently
provided only partial conversion.
sistent with (Me-IAN)AlMe2 (6a). Additionally, crystals
suitable for X-ray crystallography were obtained from
a concentrated benzene/hexanes solution (1:1). Selected
bond distances and angles are given in Figure 1.
This four-coordinate complex is necessarily distorted
from an ideal tetrahedral geometry due to the demands
of the IAN amine bite angle. The binaphthyl dihedral
angle of the bound Me-IAN ligand at 46° is substantially
decreased from those of both free Me-IAN amine (68°)
and (R-IAN)2Zr(NMe2)2 complexes (low of 58°).13 The
nonplanarity of the â-diketimine framework also ren-
ders the coordinating nitrogens electronically inequiva-
lent as expectedsa significant difference (0.1 Å) is
present in the bond distances between aluminum and
the pyridyl (1.963 Å) and naphthylamino (1.858 Å)
nitrogens. The Al-N(2) distance is 0.03 Å shorter than
the range of reported distances, and Al-N(1) is 0.03 Å
longer than any previously reported.19 The amino-
naphthalene nitrogen is essentially planar, with a sum
of its bond angles equal to 356°. In standard planar
â-diketimine systems, the two nitrogens always exhibit
equivalent bonding interactions with aluminum, or at
most a 0.01 Å difference.
When the reaction of (Me-IAN)2Zr(NMe2)2 (2a) and 2
equiv of AlMe3 was performed in a J-Young tube and
monitored by 1H NMR, peaks at 2.1 and -0.3 ppm were
observed concomitant with those ascribed to 6a, con-
sistent with the formation of Me2Zr(NMe2)2 (7). The
formation of 7 is also supported by the empirically
determined limiting stoichiometry. On the basis of these
observations, we propose that amide/alkyl metathesis
(Scheme 1, eq 1) from the initially formed Lewis base-
Lewis acid complex 3a20 is slow relative to the trans-
metalation pathway leading to 6 (Scheme 1, eq 2).
The combination of Me-IAN and trimethylaluminum
(1:1 ratio) in toluene resulted in clean conversion to a
complex whose spectral characteristics matched those
of the complex formed in the above experiments, con-
(15) (a) Kim, I.; Jordan, R. F. Macromolecules 1996, 29, 489. (b)
Diamond, G. M.; Jordan, R. F.; Petersen, J. L. J. Am. Chem. Soc. 1996,
118, 8024. (c) Thiyagarajan, B.; Jordan, R. F.; Young, V. G., Jr.
Organometallics 1999, 18, 5347.
(16) Due to the low solubility of the dihalo derivatives (R-IAN)2ZrX2,
the analogous halide/alkyl exchange could not be evaluated.
(17) For an example of a group VI (Cr(III)) â-diketiminate complex
in which chloro is converted to methyl in good yield by treatment with
AlMe3, see: Gibson, V. C.; Maddox, P. J.; Newton, C.; Redshaw, C.;
Solan, G. A.; White, A. J. P.; Williams, D. J. Chem. Commun. 1998,
1651.
(18) Gibson, V. C.; Kimberley, B. S.; White, A. J. P.; Williams, D.
J.; Howard, P. Chem. Commun. 1998, 313.
(19) (a) Cosle´dan, F.; Hitchcock, P. B.; Lappert, M. F. Chem.
Commun. 1999, 8, 705. (b) Kopp, M. R.; Neumu¨ller, B. Z. Anorg. Allg.
Chem. 1999, 625, 739. (c) Qian, B.; Ward, D. L.; Smith, M. R., III.
Organometallics 1998, 17, 3070. (d) Huang, Y.-L.; Huang, B.-H.; Ko,
B.-T.; Lin, C.-C. J. Chem. Soc., Dalton Trans. 2001, 1359. (e) Gornitzka,
H.; Stalke, D. Organometallics 1994, 13, 4398.
(20) The transmetalation occurs so rapidly that this complex is not
observed here, but its formation as an intermediate is presumed on
the basis of Jordan’s work.6
Figure 1. Thermal ellipsoid (50%) plot of 6a. H atoms are
removed for clarity. Selected bond distances (Å) and angles
(deg): Al-N(1) ) 1.9633(17), Al-N(2) ) 1.8582(18), Al-
C(21) ) 1.961(2); N(1)-Al-N(2) ) 92.95(7), N(1)-Al-C(21)
) 107.96(8), N(1)-Al-C(22) ) 108.17(9), N(2)-Al-C(21)
) 113.94(9), N(2)-Al-C(22) ) 112.25(9), C(21)-Al-C(22)
) 118.26(10).

Communications
Organometallics, Vol. 23, No. 25, 2004 5887
Table 1. Ethylene Polymerization Activity of IAN Amine Aluminum(III) and Zironium (IV) Precatalystsa
amt of cat
(µmol)
amt of activator
efficiency
(g of PE/g of Al)
entry
precatalyst
activator
(µmol)
activity
1
2
3
4
5
6
(Bn-IAN)AlMe2 (6b)
(Bn-IAN)AlMe2 (6b)
(Bn-IAN)AlMe2 (6b)
(Me-IAN)AlMe2 (6a)
(Bn-IAN)AlMe2 (6c)
(Ph-IAN)2Zr(NMe2)2 (2c)
9.7
9.7
4.0
4.1
4.0
5.0
PhNMe2HBArf
Ph3CBArf
MMAO
5.6
8.7
2000
4800
2000
1000
18
609
3100
780
3250
780
662
22580
114900
28928
120460
8551b
MMAO
MMAO
MMAO
i
a All reactions were performed with 1 L batch reactors using 2000 µmol of Bu3Al scavenger at 70 °C (85 °C for entry 6) for 0.5 h. The
ethylene pressure was 200 psi. Activity is given in units of g mmol-1 h 100 psi-1 b Efficiency in g of PE/g of Zr.
.
However, complex 7 appears volatile, as it is consist-
ently removed under vacuumsnot unlike the method
for removal of aluminum complex 5.
IAN)AlMe2 as a representative example, activation with
anilinium-BArf salt provided a catalyst with only low
activity. Use of the more potent activator trityl-BArf
salt provided a substantially more active catalyst (Table
1, entry 2).27 Gratifyingly, MMAO further increased the
activity by 6-fold. Under otherwise identical conditions,
(Me-IAN)AlMe2, (Bn-IAN)AlMe2, and (Ph-IAN)AlMe2
produced catalysts with increasing activity, correspond-
ing to efficiencies of 28 928, 114 900, and 120 460 g of
PE/g of Al, respectively (Table 1, entries 4, 3, and 5).
Direct activation of (Ph-IAN)2Zr(NMe2)2, a complex that
readily transmetalates in the presence of AlMe3, did give
rise to an active catalyst (Table 1, entry 6); however,
the comparison in Table 1 clearly shows a difference
relative to the corresponding aluminum complex (Table
1, entry 5). This result can be interpreted in two ways:
(1) the polymerization activity for (Ph-IAN)2Zr(NMe2)2
is derived from a catalyst of general structure (Ph-
IAN)2ZrX2 that is only weakly active or (2) the polyeth-
ylene is produced by a small amount of (Ph-IAN)AlX2
that is moderately active, as demonstrated in Table 1,
entry 5.28 Despite this ambiguity, this represents to our
knowledge the first case in which the aluminum com-
plex was found to be a more reactive olefin polymeri-
zation precatalyst than its zirconium counterpart.29 This
may in part be due to the potential reactivity of the
carbon backbone of typical aluminum â-diketiminates
for which side reactions with ethylene have been
observed.29 Our current hypothesis is that the electron-
ics of the IAN-amine ligand (â-amido Schiff base)
provides an opportunity unique among typical planar,
electronically delocalized â-diketimines to slow â-
hydride elimination and/or ligand incorporation of eth-
ylene.
That the transmetalation is so rapid is counter-
intuitive when considering the substantially lower
Lewis basicity of the naphthylamino nitrogen relative
to the dimethylamino nitrogen.21 Moreover, the naph-
thylamino nitrogen is considerably hindered, on the
basis of kinetic studies of R-IAN atropisomerization.11
To test the notion that the transmetalation may be
simply a pathway driven by formation of 6 as a
thermodynamic sink, a series of attempts to achieve
kinetic selectivity for the dimethylamido nitrogens was
made. However, addition of AlMe3 to 2a at temperatures
as low as -78 °C led to formation of the same product
as that observed at room temperature.
Further evidence that this behavior was not simply
due to the hindered nature of the 2:1 complexes in which
initially formed adduct 3a sterically prohibited methyl
transfer and release of the first dimethylamido ligand
was gathered by combining (Me-IAN)Zr(NMe2)3(HNMe2)
and AlMe3 (6 equiv) in an attempt to form the tri-
alkylzirconium complex (Me-IAN)ZrMe3. This reaction
also furnished the product of transmetalation. These
observations extend to the analogous Bn-IAN (2b) and
Ph-IAN (2c) complexes.
Atypical behavior has been observed in metathesis
strategies applied to mixed alkoxide/Schiff base
ligands that form an unhindered, relatively strained
five-membered chelate.22-24 Also, tetradentate bis(â-
diketimine) complexes of group IV metals typically
result in the product of interrupted amido transfer.6
This behavior is uncharacteristic of bidentate zirconium
amido complexes as well. For example, 1,8-bis((tri-
methylsilyl)amino)naphthalene readily complexes to
zirconium(IV) to form a complex that is unreactive
toward AlMe3, despite its rapid direct reaction with
AlMe3 to form the expected bis(aluminum) complex.25
The dihalozirconium complexes of chelating silyl bis-
(amides) also readily exchange chloride with methyl
upon exposure to AlMe3.26
We have also considered the possibility that the
counterion formed upon activation of a zirconium di-
methylamide may render the catalyst less reactive than
that formed from the corresponding aluminum alkyl.
However, we have previously determined that L2Zr-
(NMe2)2 and L2ZrCl2 (L ) IAN-amine) exhibit compa-
rable reactivity,13 and in general, the variability of
activity in counterion studies is typically limited to less
than a factor of 10. Moreover, aluminum complexes are
Ethylene Polymerization. Table 1 describes initial
ethylene polymerization results for a representative
group of (R-IAN)AlMe2 complexes (6a-c). Using (Bn-
(27) Polymerizations are normally done in concert with activator-
only runs, as well as standard (highly active) activators that provide
the proper controls.
(28) Use of polymer molecular weight or PDI to determine whether
the catalyst is aluminum or zirconium is not possible, since zirconium
complexes produce such a variety of polymers.
(21) For an example of partial transmetalation of a bis(amide), see:
Ison, E. A.; Abboud, K. A.; Ghiviriga, I.; Boncella, J. M. Organometallics
2004, 23, 929.
(22) Bei, X.; Swenson, D. C.; Jordan, R. F. Organometallics 1997,
16, 3282.
(23) Coles, M. P.; Hitchcock, P. B. Dalton 2001, 1169.
(24) (a) Tsukahara, T.; Swenson, D. C.; Jordan, R. F. Organo-
metallics 1997, 16, 3303. (b) Kim, I.; Nishihara, Y.; Jordan, R. F.
Organometallics 1997, 16, 3314.
(29) For examples of cationic three-coordinate aluminum complexes
that undergo â-hydride elimination and reversible ethylene addition
to the anionic ligand, see: (a) Radzewich, C. E.; Coles, M. P.; Jordan,
R. F. J. Am. Chem. Soc. 1998, 120, 9384. (b) Korolev, A. V.; Ihara, E.;
Guzei, I. A.; Young, V. G., Jr.; Jordan, R. F. J. Am. Chem. Soc. 2001,
123, 8291. Calculations: (c) Talarico, G.; Busico, V.; Budzelaar, P. H.
M. Organometallics 2001, 20, 4721.
(25) Lee, C. H.; La, Y.-H.; Park, S. J.; Park, J. W. Organometallics
1998, 17, 3648.
(26) Hill. M. S.; Hitchcock, P. B. Organometallics 2002, 21, 3258.

5888 Organometallics, Vol. 23, No. 25, 2004
Communications
typically multiple orders of magnitude less reactive than
their zirconium counterparts. Although a direct com-
parison cannot be made for the IAN-amine ligands and
other aluminum â-diketimines, these systems compare
favorably to the monoanionic nitrogen-based aluminum
systems previously reported.30
general for planar and/or hindered â-diketimines, we
have established it in the context of a â-diketimine that
binds at the extreme as a â-amido Schiff base. Future
studies are aimed at further clarification of this guide-
line and a more precise weighting of the roles of both
steric and electronic effects on transmetalation.
In summary, complexes of the type (R-IAN)2ZrNMe2
readily transmetalate to aluminum to give (R-IAN)-
AlMe2, and these complexes exhibited substantial activ-
ity in the polymerization of ethylene. This behavior is
counterintuitive on the basis of an analysis of the Lewis
basicity of the three nitrogen types present in 2a and
underscores the care one must apply when interpreting
polymerization activity of catalysts bearing amido-type
ligands with aluminum activating agents. Although it
remains to be determined whether this behavior is more
Acknowledgment. This work was supported by
Indiana University and a Yamanouchi Faculty grant.
Acknowledgment is made to the GAANN program
(S.B.C.) for predoctoral fellowship support. We also
thank Dow technologists Mike Allen and Gorden O’Dell
for their contributions to the polymerization studies.
Supporting Information Available: Text giving general
experimental procedures and spectral data for all new com-
pounds, and a CIF file giving complete X-ray structural data.
This material is available free of charge via the Internet at
http://pubs.acs.org.
(30) (a) Ihara, E.; Young, V. G., Jr.; Jordan, R. F. J. Am. Chem. Soc.
1998, 120, 8277. (b) Coles, M. P.; Jordan, R. F. J. Am. Chem. Soc. 1997,
119, 8125.
OM049364G